7916
Although it is quite clear that in 96% H2S04isomerization occurs via a T,T* excited state of 2a, the nature of the excited states of neutral 4-pyrones has not been established. It seems likely, however, that the similar photoreactivity of l a and 2a also reflects a similarity of their respective excited states. In order to test such a suggestion we have irradiated neutral 4-pyrones l a and l b in 2,2,2-trifluoroethanoI, a highly polar solvent that would be expected to stabilize T,T* excited states relative to n,T* excited states. l 5 The relative quantum efficiencies of photoisomerization of l a and l b in acetonitrile and trifluoroethanol tend to support the suggestion that the isomerization of neutral 4-pyrones also occurs via their T , T * excited states. Thus irradiation of equimolar solutions of l a in acetonitrile and in trifluoroethanol in a merry-goround apparatus resulted in a 30 greater conversion of l a in the latter, more polar solvent. This dependence on solvent polarity was particularly notable in the case of l b . Although photostable in a~etonitrile,~we observe that in trifluoroethanol l b is converted to 3b. These results are inconsistent with the suggestion that the efficiency of the photoisomerization is sterically controlled by the large phenyl groups at positions 3 and 5 of the 4-pyrone ring.4 Indeed, our observation that tetramethyl-4-pyrone (IC) undergoes efficient photoisomerization to 3c in either acetonitrile or trifluoroethanol clearly shows that phenyl groups in these positions are not a prerequisite to photoisomerization. Rather, these results suggest that the nature of the excited states and hence the reactivity of 4-pyrones is sensitive to the extent of substitution in the 4-pyrone ring. Acknowledgment is made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for support of this research. (15) J. Griffiths and H. Hart, J. Amer. Chem. Soc., 90, 5296 (1968). (16) ACS-PRF Undergraduate Research Participant, 1972-1973. James W. Pavlik," Josephine Kwong16 Driparrmerit of Chemistry, Utiicersity of Wiscorisin-Riaer FaNs Ricer FaNA, Wiscorisiri 54022 Receiwd July I I , 1973
Isotope Effects in the Solvolysis of Haloallenes' Sir: We wish to report the first measurement of an a secondary isotope effect for an sp2 to sp hybridization change in a solvolysis reaction. We have been examining the behavior of tri- and disubstituted haloallenes under solvolytic conditions. 2 As a mechanistic probe we have measured the a- and 6-secondary isotope effects associated with the solvolysis of la-e in aqueous ethanol and trifluoroethanol. These data are collected in Table I. Data derived from Table I support the proposal that these compounds all react via a carbonium ion mechanism. The enhanced reactivity of these haloallenes (1) Submitted in partial fulfillment for the degree of Master of Arts in Chemistry, College of William and Mary, 1973. (2) (a) M. D. Schiavelli, S . C. Hixon, H. W. Moran, and C. J. Boswell, J . Amer. Chem. SOC.,93, 6989 (1971); (b) M. D. Schiavelli, R. P. Gilbert, W. A. Boynton, and C. J. Boswell, ibid., 94, 5061 (1972); (c) M. D. Schiavelli and P. L. Timpanaro, J . Org. Chem., 38,3054 (1973). (3) It is understood throughout that ion pairs are the most likely candidates for intermediates in these solvolyses. Preliminary results appear to confirm this.
Journul of the Americun Chemical Society
/
Br
Table 1. Conductometric Rate Constants for Solvolvsis of 1 Compd
T , "C
Solventa
la
24.62 f 0 . 0 1
60E 70E 60E 70E 8OE 80E 90E 60E 70E 80E 90E 70T
34.68 i 0.01 45.30 f 0.02
lb
24.62 i 0.01 45.30 i 0 . 0 2
IC Id le
50.08 f 0 . 0 1 60.22 f 0.01 70.13 f 0.01 60.24 f 0.01 60.22 f 0.01
IO4k, sec-lb 3.64 i 0 . 0 6 1 . 1 4 f 0.02 12.1 f 0 . 1 4.28 i 0.03 1.47 f 0.02 5.15 f 0 . 0 4 1.49 f 0 . 0 3 3.10 i 0 . 0 4 0.98 f 0.004 4.31 f 0.04 1.22 f 0.07 0.674 i 0.003 2.44 i 0 . 0 2 5.21 f 0 . 0 1 2 . 0 5 =k 0.002 2.03 i 0 . 0 1
70T 70T
60E represents 60 : 40 (v/v) ethanol-water, etc. ; 70T represents 70: 30 (w/w) trifluoroethanol-water. * Average of triplicate determinations.
Table 11. Isotope Effects for Solvolysis of 1
la/lb
lc/ld lc/le
24.62
60E 70E
45.30
80E
60.2 60.2
90E 70T 70T
1.17 1.16 1.19 1.22 1.19 1.20
37 35 44 50 44 129
1.21 0.03 1.19 i 0.02 1.23 i 0.02 1.27 i 0 . 0 4 1 . 2 3 f 0.01 1.22 i 0.01
Corrected to 100% deuteration at the temperature of the measurement. (The authors are indebted to Professor Donald Hunt, University of Virginia, for the mass spectral analysis of Calculated from A A F S . lb, d, and e.) Q
compared to their vinyl analogs allows the convenient measurement of these rates even though the developing cation is not stabilized by a substituent at the reaction center in IC, d, and e . 4 The solvent dependence of l a yields m = 0.72 + 0.05 at 45'. The temperature dependence data yield the following activation parameters: for l a in 70E AH* = 22.1 kcal/mol and AS*(25') = -2.3 eu, and for I C in 70T AH* = 21.9 kcal/mol and AS*(25') = -9.9 eu. Table I1 presents the isotope effects observed. The 6-secondary isotope effect observed upon CDs substitution at the 3 position of a trisubstituted haloallene (la/lb) is found to be kH/kD= 1.23 (44 cal/D) in 80% ethanol at 45". This is to be compared with that reported by Shiner and Kriz for solvolysis of 2 where C1
c1
I
(CH3)zC-CsCCH
I
3 US.
(CD &C-C=CCH
3
2
(4) It should be noted that these haloallenes exhibit relative reactivity about one order of magnitude lower than fert-butyl chloride and CYphenylethyl chloride and are about as reactive as benzyl halides.
1 95:23 1 November 14, 1973
7917
k H / k D= 1.65 (50 ~ a l / D ) . ~In the case of the terminal bromoallene, Id, kH/kD= 1.23 (44 cal/D) for CDs substitution at C-3 in 70T at 60”. These data strongly support the proposal that terminal haloallenes as well as trisubstituted haloallenes react via charge-delocalized cationic intermediates under initially neutral conditions. The a-secondary isotope effect observed upon deuterium substitution at the reaction center kH/kD = 1.22 represents one of the largest a-secondary isotope effects yet reported. For saturated chlorides, such as 1phenylethyl chloride, Shiner has suggested that values of kH/kn = 1.15 (83 cal/D) are to be considered a maximum for limiting solvolysis.6 Shiner has also noted that these ”limiting” maxima vary with the leaving group. Thus, the a effect reported here is to be compared with that for bromide as the leaving group, k H / k D= 1.125 (70 ~ a l / D ) . ~Larger a effects are observed in the solvolysis of secondary propargyl brosylates6 and furylmethylcarbinyl p-nitrobenzoate.8 Shiner has suggested that kHlkD 1.23 for oxygen leaving groups in saturated systems be considered comparable to the lower maximum effects observed with the halides.’ We are unable to assess the magnitude of the limiting a effect to be expected for an sp2 + sp hybridization change from the data reported here. However, it is possible to estimate this maximum from a consideration of the exchange equilibrium constants for the reactions shown below.9 These calculations imply that the isoCHaCH,
+ CH-CHD
I_ CHaCH2D + CH%=CHZ K
CH2=C€I2
=
1.075
+ D C z C H I_ CHz=CHD + HC=CH K = 1.257
tope effect for sp2 + sp hybridization is 1.257/1.075 = 1.17 times that for an sp3 + spz change. Assuming a maximum kH/kD = 1.125 for Br- leaving from a saturated halide, this predicts that the maximum cu-isotope effect to be expected for solvolysis of a vinyl or allenyl bromide is 1.32,’O Thus it appears that larger a-isotope effects may be observed for systems on which digonal cations are implicated. We are continuing work in this area by varying solvent, temperature, and structure in order to assess the degree of limiting character in these solvolyses as well as to test the prediction that digonal cations assume a linear geometry.”
of the College of William and Mary Faculty Research Fellowship Program. Melvyn D. Schiavelli,* D. Ellen Ellis Department of Chemistry, Coffegeof‘ WiNiam and M a r y Wilfiamsburg, Virginia 23185 Receiced M a y 31, 1973
Selenium Dioxide Oxidations of Olefins. Trapping of the Allylic Seleninic Acid Intermediate as a Seleninolactone Sir : A recently proposed mechanism (Scheme I) for the Scheme I
X. Y = OH, 0.4~.01’ OR
oxidation of olefins by selenium dioxide suggested an initial ene addition of an >Se+-0- moiety (step a) followed by dehydration, or its equivalent (step b), and a [2,3] sigmatropic shift (step c) of the resulting allylseleninic acid.’ Having previously’ provided support for the [2,3] shift, we now report evidence in favor of the initial ene reaction by demonstrating that with appropriate substrates, such as 2 and 6, the intermediates can be trapped as seleninolactones (Scheme 11). Scheme I1
COOCH, Se02 benzene
Z.R=H
COOCH
3.R=k 4, R = &O,H
05.34%
Acknowledgments. Acknowledgment is made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for support of this research. The senior author acknowledges the support
FH
(5) V. J. Shiner, Jr., and G. S. Kriz, Jr., J . Amer. Chem. SOC.,86, 2641 _ . (1964). ~~. . ,_
(6) V. J. Shiner, Jr., in “Isotope Effects in Chemical Reactions,” C. J. Collins and N. S. Bowman, Ed,, ACS Monograph 167, Van NostrandReinhold, New York, N. Y., 1970, pp 105-120. (7) V. J. Shiner, Jr., W. E. Buddenbaum, B. L. Murr, and G. Lamaty, J . Amer. Chem. SOC.,90,418 (1968). ( 8 ) D . S. Noyce and G. V . Kaiser, J . Org. Chem., 34,1008 (1969). (9) S. R. Hartshorn and V. J. Shiner, Jr., J. Amer. Chem. SOC.,94, 9002 (1972). (IO) In fact, a more appropriate exchange equilibrium constant required for prediction of this maximum is that for the reaction CH?=C= CHBr DCECH CHz=C=CDBr H C s C H . The calculations on this system are being attempted. (11) J. E. Williams, R. Sustmann, L. C. Allen, and P. v. R. Schleyer, J . Amer. Chem. SOC.,91, 1037 (1969).
+
+
R O / Y
t-BuOH
+&A
0’
8.395 6,R=H 7, R = &O,H
Treatment of sordaricin methyl ester 2 2 , 3with sele(1) I
